Investor attention and Google Search Volume Index: Evidence from an emerging market using quantile regression analysis

Abstract

This study investigates whether the investor attention measured by the Google Search Volume Index (GSVI) is effective in forecasting stock returns. The evolving literature on investor attention suggests that higher GSVI can predict higher returns for the first one or two weeks, but with a subsequent price reversal. We use a more recent dataset that covers S&P BSE 500 companies listed on the Indian stock exchange for 2012–2017 and employ the quantile regression approach because it alleviates the statistical problems arising from biased distribution data. The results suggest that a higher GSVI predicts positive and significant returns in the subsequent first and second weeks. Higher quantiles of GSVI experience higher excess returns. The panel cointegration test results support the findings regarding the cointegration of the GSVI and stock returns. Our empirical evidence shows that our model is robust when using a trading strategy based on the Fama-French four-factor model. Thus, the model with GSVI acts as a better predictor of both the direction and magnitude of the excess returns than the model without GSVI.